The goal is to compare implicit and explicit memory in healthy aged and aged with probable Alzheimer's disease (AD) where implicit and explicit memory tasks share a common paradigm and a common metric based on selective attention to visual novelty. The selective attention test consists of a series of "novelty problems." For each problem in the test an individual studies a picture for a specific period of study time then that picture is paired with a new picture. To test implicit memory, participants are simply instructed to "look at the pictures." Typically, more visual fixation is paid to the novel target when the two targets are paired. To test explicit memory, participants are told to "look at the new picture". In the present study, we will explore the influence of instructions, exposure duration of pictures, and retention interval on tests of selective attention in comparisons of normal elderly and elderly with probable AD. Theoretically, we will discover if dissociations between implicit and explicit memory due to illness, length of study, and retention interval hold when both kinds of memory are measured in the same way. The practical implications of the study rest on the fact that selective attention to novelty is predictive of a person's overall cognitive functioning within samples of infants, young adults, closed head injury patients, normal elderly, and elderly with probable AD. The link between differential visual fixation to novelty and cognitive functioning provides a means for measuring cognition in individuals such as elderly with probable AD for whom conventional tests of cognitive function may be inappropriate. Knowledge of the parameters constraining selective attention will enhance the utility of selective attention testing for initial cognitive screening for demented individuals and also for longitudinal assessment of cognitive functioning over the course of illness.